Simulation study as supplement to EEMS proposal 2017-0029, “Circulating Estrogens, Androgens, and Progestogens and Risk of Breast Cancer in the PLCO Cohort”.
The simulation will require the following:
A SAS file with data for all women potentially eligible for the hormone case-control studies, including:
1) All cases (N= 277) and controls (N = 423) included in the phase I hormone case-control study with indicator variables identifying them.
2) Additional breast cancer cases diagnosed prior to 6/30/2005 that were not included in phase I of the study but will be included in the expanded study (e.g. those with serum from other than baseline, those with other cancers) and an indicator variable to identify them.
3) All new breast cancer cases diagnosed since 6/30/2005 to be included in the expanded analysis with an indicator variable to identify them.
4) Those who died or were lost to follow-up by 6/30/2005 and who would have otherwise met the criteria for inclusion in the hormone case-control study. Create an indicator variable to identify them.
5) The potential controls who had not died or were lost to follow-up by 6/30/2005 and who met eligibility criteria for the hormone phase I case-control study as of baseline, but were not selected. Create indicator variables to identify this group.
Variables to include in addition to indicator variables described above:
For subjects included in case-control analysis
1) Unconjugated estradiol values for case and controls included in phase I case-control analysis.
For all individuals
2) Variables used for frequency matching (age at study entry and date of blood draw)
3) Age at breast cancer diagnosis, death, lost-to-follow-up, or exit from study,
age at menarche, combined parity and age at first birth, age at natural menopause, type of menopause, first degree family history of breast cancer, personal history of benign breast disease, previous use of menopausal hormone therapy, body mass index (BMI).
Mitchell Gail
Catherine Schairer
Regina Ziegler